%B Planning, complexity and new ICT %L eprints1887 %D 2009 %X We study the patterns of the communities of workers and students of Sardinia, by applying grouping methodologies based on the characterization of the Sardinian commuting system as a complex weighted network. The algorithm is based on a heuristic method able to optimize a quality function called modularity, as proposed by Newman et al. [01]. Its adoption allows us to detect productive basins geographically located, as composed by towns and their territories showing a certain degree of similarity. We compare the spatial distributions of communities with relevant historical and provincial boundaries, investigate on their discrepancies and correspondences, and propose possible perspectives for local policy making and planning. %A Simone Caschili %A Andrea De Montis %A Alessandro Chessa %A Giancarlo Deplano %K Complex networks, Community detection, Commuting networks, Weighted networks, Spatial networks %I (ed. by G. Rabino, M. Caglioni) Planning, complexity and new ICT %T Weighted networks and community detection: planning productive districts in Sardinia %P 27-36